© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 32-TRANSCODER-HP
    7. 43027
    Prev
    Next
    INDEX
    Explanations

    Graves and cemeteries

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    atement
    -0.28
    enter
    -0.26
    å²Ķ
    -0.25
    å¡
    -0.25
    æĢĢ
    -0.25
    ãĤĴæ±ĤãĤģ
    -0.25
    å£ij
    -0.24
    女åĦ¿
    -0.24
    æŀľ
    -0.24
    quito
    -0.24
    POSITIVE LOGITS
    津贴
    0.29
    ndx
    0.27
    indexes
    0.26
    çĶĺ
    0.26
     indexes
    0.26
    lineno
    0.26
    maps
    0.25
    çªĺ
    0.25
    boards
    0.25
    .exc
    0.25
    Activations Density 0.001%

    No Known Activations